Abstract: The Human Microbiome Project (HMP) is generating a wealth of data about the composition and dynamics of human-associated microbial communities, and other recent efforts have shown that changes in the composition of the microbiome are correlated with human diseases. This proposal, which originated with my laboratory's recent and unexpected finding that human-associated bacteria produce drug-like small molecules, will use HMP data as a starting point for identifying and characterizing new small molecules from human-associated microbes. There are three convergent motivations for this project: First, the HMP, an NIH Roadmap Initiative, is generating a wealth of data about the composition and dynamics of human-associated microbial communities. New connections between the composition of the microbiome and human disease are being discovered with increasing frequency, but functional studies are needed to translate this data into new therapies. This project will build one such bridge, using HMP data to support functional studies on human-associated bacteria that could lead to new probiotic therapies. Second, small molecules from human-associated bacteria are likely to have important activities. Given the strong precedent for microbially-produced small molecules to mediate interspecies interactions, small molecules from the human microbiome are likely to mediate intra-microbiome and host- microbiome interactions, both of which will have an important bearing on human health. This project will use HMP data to identify and characterize new small molecules from the human microbiome that mediate interspecies interactions. Third, engineered probiotics are likely to be common in the future, and will be one of the most direct benefits to human health from the Human Microbiome Project. Unlike the human genome, the human microbiome is a component of our 'supra-genome'whose composition we can control. This project will lay the groundwork for engineering new probiotic bacterial strains, creating the potential for a direct link from the HMP to the clinic. Public Health Relevance: Natural products - small molecules from microbes - have been a primary source of clinically-used antibiotics, antifungals, anticancer agents, immunosuppressants, and other drugs. This proposal describes new approaches to discovering and characterizing natural products from an unusual source: human-associated microbes. Identifying and characterizing these molecules will not only lead to a greater understanding of the connection between the human microbiome and human disease, it will also lay the groundwork for engineering new probiotic bacterial strains (like the bacteria found in yogurt) that could promote human health and help fight diseases like Crohn's disease and obesity.